Software for generating synthetic astronomical observables
Show detailsHide details
Synthesizer is our main forward-modelling software effort. It is a modular Python package for generating synthetic astrophysical observables, designed to be fast and extensible. It supports direct comparisons between simulations and data in observational space, and is now used in multiple recent studies.
Predicting the most extreme objects in the Universe
Show detailsHide details
This project applies extreme value statistics to predict the most massive haloes and galaxies at high redshift. It was used to quantify early JWST tension claims and how those changed with improved calibration, redshift vetting, and stellar-mass estimates.
I use FLARES and related simulations to study when the first passive galaxies emerge, and what physical mechanisms shut down star formation in the early Universe. The work links rare high-redshift passive candidates to AGN feedback pathways in the models.
This line of work uses machine learning methods to accelerate and extend simulation-based science, including galaxy-halo mapping and probabilistic modelling. It also includes applications to spectra, cosmological inference, and uncertainty-aware predictions.
Simulation-based inference combines forward models with modern neural density estimation for likelihood-free Bayesian analysis. I apply these methods to astrophysical parameter inference and model comparison, including work with LtU-ILI and synthetic CAMELS photometry.
Modelling stellar populations, for observational and theoretical applications
Show detailsHide details
Stellar population synthesis is a core ingredient in many of my forward models. This work spans model development, practical tools such as Sengi, and applications to nebular line and continuum predictions with collaborators.
Here I study dusty star-forming galaxies in the early Universe and their far-IR to sub-mm observables. Using hydrodynamic simulations and dust radiative transfer, this project explores number counts, orientation effects, and survey-scale predictions.
This project focuses on identifying and characterizing the progenitors of present-day massive galaxy clusters before collapse. It includes work on detection strategies, morphology, and broader environmental effects on galaxy evolution.